Intercalated cells of the renal collecting tubule are specialized for H+ and HCO3-transport. We had found that these cells exist in two forms; the alpha type has a H+ ATPase in the apical membrane and a Cl/HCO3 exchanger in the basolateral membrane. The beta type has these proteins reversed in polarity. We previously demonstrated that feeding animals an acid diet reversed the polarity of the beta cells to that of the alpha type. We have now produced a clonal immortalized beta cell line and demonstrated plating them at high density converts the cells to the alpha form in vitro. High density induces the production of a 230 kDa extracellular matrix protein which is sufficient by itself to produce the beta -> alpha conversion. We plan to purify this protein to homogeneity, examine its molecular structure, its receptor and the mechanism by which the binding to its receptor could induce this conversion. In the clonal cell line we also found that the Cl/HCO3 exchanger of both the alpha and beta type of cells is the same protein, band 3. We aim to examine its association to various components of the actin based cytoskeleton such as ankyrin, fodrin and protein 4.1 in alpha and beta cells. We also plan to examine the targeting pathway for newly synthesized band 3 to opposite cell domains under the influence of the extracellular matrix protein. Epithelial polarity is not only determined by external cues such as the extracellular matrix, rather these cells must have important regulatory genes that allow them to express hundreds of epithelial proteins and to repress as many non-epithelial proteins. To identify this gene we reasoned that it must be expressed in cells that have just become epithelialized such as occurs during nephrogenesis. We generated an induced kidney mesenchymal cell line that expresses four epithelial genes (uvomorulin, type IV collagen, laminin and desmoplakin) as well as having retained some mesenchymal proteins. A cDNA library from this cell was transfected into fibroblasts. We isolated primary transfectant clones that have now started to express epithelial genes and to repress mesenchymal. This gene when fully isolated should allow us to identify the genetic basis of epithelial determination.